1. Volume 48, Issue 4, Pages 760-772.e4 (April 2018)
Interleukin-15 Complex Treatment Protects Mice from Cerebral Malaria by Inducing Interleukin-10-Producing Natural Killer Cells 
Kristina S. Burrack, Matthew A. Huggins, Emily Taras, Philip Dougherty, Christine M. Henzler, Rendong Yang, Sarah Alter, Emily K. Jeng, Hing C. Wong, Martin Felices, Frank Cichocki, Jeffrey S. Miller, Geoffrey T. Hart, Aaron J. Johnson, Stephen C. Jameson, Sara E. Hamilton 
Immunity 
Volume 48, Issue 4, Pages e4 (April 2018)
DOI: /j.immuni
Copyright © 2018 Elsevier Inc. Terms and Conditions

2. Immunity 2018 48, 760-772.e4DOI: (10.1016/j.immuni.2018.03.012)
Copyright © 2018 Elsevier Inc. Terms and Conditions

3. Figure 1 IL-15C Prevents the Development of ECM
(A) Survival curve of mice left untreated (n = 15) or treated prophylactically with IL-2C (n = 14) on days −7, −5, and −3 or IL-15C (n = 13) on days −5 and −3 relative to PbA infection. Data are combined from three independent experiments. ∗∗∗∗p < as determined by log-rank (Mantel-Cox) test.
(B) Survival curve of mice left untreated (n = 14) or treated therapeutically with IL-15C (n = 20) on days 2 and 3 after PbA infection. Data are combined from four independent experiments. ∗∗∗∗p < as determined by log-rank (Mantel-Cox) test.
(C) Parasitemia analysis on day 6 pi of mice left untreated or treated prophylactically or therapeutically with IL-15C. Each symbol represents one mouse. Data are combined from three independent experiments. Each symbol represents one mouse; horizontal bar indicates the mean. ∗p < 0.05 as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
(D) PbA-infected mice (n = 7/group) were treated therapeutically with PBS or IL-15C. Representative gadolinium-enhanced T1-weighted MRI scans (midsagittal orientation) of mice on day 6 pi. Arrowheads indicate regions of hyperintensity.
(E) Quantification of hyperintensity measured from T2-weighted MRI of the same mice as in (D). Uninfected mice (n = 5) were used as a control. Each symbol represents one mouse; data are presented as the mean ± SEM. Data are combined from two independent experiments. ∗∗p < 0.01 as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
(F) PbA-infected mice (n = 5/group) were treated therapeutically with PBS or IL-15C on days 2+3 pi. Representative images of coronal slices from cerebral cortex showing FITC-albumin leakage (green) with DAPI counter-stain (blue).
(G) Quantification of cerebral FITC leakage from mice in (F). Uninfected mice (n = 4) were used as a control. Each symbol represents one mouse; data are presented as the mean ± SEM. ∗∗∗p < 0.001, ∗∗∗∗p < as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
Related to Figure S1.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

4. Figure 2 IL-15C-Treated NK Cells Prevent the Development of ECM
(A and B) Total number of CD8+ T cells (A) and NK cells (B) in the spleens of mice left untreated (n = 13) or treated with IL-2C three times (n = 17) or IL-15C two times (n = 16). Data are combined from six independent experiments and are presented as the means ± SEM.
(C) Mice were treated with IL-2C on days −7, −5, and −3 and IL-15C on days −5 and −3 relative to LM-OVA infection. CFU of LM in the spleen on day 5 pi. Each symbol represents one mouse; horizontal bar indicates the mean. Data are combined from two independent experiments. LOD, limit of detection.
(D) Survival curve of mice (n = 9/group) left untreated or treated prophylactically with IL-2C on days −7, −5, and −3 or IL-15C on days −5 and −3 and/or treated with anti-NK1.1 Ab on days −3, −1, and +1 relative to PbA infection. Data are combined from two independent experiments.
(E) Survival curve of mice that, 1 day prior to PbA infection, received NK cells enriched from the spleens of mice left untreated or treated with IL-2C three times or IL-15C two times. Groups: PbA (n = 15), NK + PbA (n = 7), IL-2C NK + PbA (n = 10), IL-15C + PbA (n = 15). Data are combined from three independent experiments.
(A–C) ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p <  as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
(D and E) ∗∗∗p < 0.001, ∗∗∗∗p < , as determined by log-rank (Mantel-Cox) test.
Related to Figure S2.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

5. Figure 3
IL-15C Treatment Has Little Effect on T Cell Numbers in the Spleen and Brain
The spleens and brains of untreated or prophylactically IL-15C-treated PbA-infected mice (n = 9/group) were harvested on day 6 pi for CD8+ T cell analysis. Gating strategy to identify GAP50-specific CD8+ T cells in the spleen (A) and brain (D). The frequency (B and E) and number (C and F) of total and GAP50-specific CD8+ T cells in the spleen (B and C) and brain (E and F) were quantified. Data are combined from three independent experiments and presented as the mean ± SEM (B, C, E, F). ∗p < 0.05, ∗∗p < 0.01, ∗∗p < 0.01, ∗∗∗∗p < as determined by two-way Student’s t test.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

6. Figure 4
IL-15C Treatment Results in Decreased T Cell Activation in the Brain on Day 6 pi
The brains of PbA-infected Nr4a1(Nur77)GFP (n = 9/group) or IfngYFP mice (n = 6–7/group) that were left untreated or treated with IL-15C on days −5 and −3 relative to PbA infection were harvested on day 6 pi. Infiltrating leukocytes were isolated and analyzed by flow cytometry. Representative histograms of Nr4a1GFP or IfngYFP expression in total CD8 T cells or GAP50-specific CD8+ T cells from Nr4a1GFP (A) or IfngYFP (E) mice. To combine multiple experiments, the Nr4a1GFP (B) or IfngYFP (F) MFI was normalized and graphed. The frequency (C and G) and total number (D and H) of Nr4a1GFP+ (C and D) or IfngYFP+ (G and H) CD8 T cells were quantified. Data are combined from three (A–D) or two (E–H) independent experiments and presented as the mean ± SEM (B–D, F–H). ∗p < 0.05, ∗∗p < 0.01 as determined by one-way ANOVA followed by Tukey’s multiple comparison test. Related to Figure S3.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

7. Figure 5 IL-15C Treatments Induce IL-10 Expression in NK Cells
(A) WT mice (n = 3/group) were treated with IL-2C on days −7, −5, and −3 or IL-15C on days −5 and −3 relative to harvest, and splenic NK1.1+NKp46+CD3− NK cells were FACS sorted. RNA-seq analysis was performed on RNA isolated from the purified NK cell populations. Genes shown were significantly differentially expressed between the two treatments with ≥2-fold change in expression. Arrowhead indicates Il10 on the heatmap.
(B and C) Il10GFP mice were left untreated (n = 11) or treated with IL-2C (n = 8) or IL-15C (n = 11) on days −5 and −3 relative to harvest.
(B) Representative histograms of Il10GFP expression in NK1.1+NKp46+TCRβ− cells in the spleen, blood, and brain.
(C and D) The frequency (C) and total number (D) of Il10GFP+ NK cells in the spleen, blood, and brain on days 0, 3, and 6 after PbA infection were quantified. Data are combined from four independent experiments and presented as the mean ± SEM (C and D); ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
Related to Figures S4 and S5.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

8. Figure 6
Human IL-15C Treatment Induces IL-10 Expression in NK Cells and Protects Mice against ECM
Il10GFP mice were left untreated (n = 5) or treated with IL-15C (n = 4) or ALT-803 (n = 5) on days −5 and −3 relative to harvest.
(A–D) The frequency (A) and number (B) of splenic NK cells were quantified. After gating on NK cells, the frequency (C) and number (D) of Il10GFP+ cells were quantified. Data are combined from two independent experiments and presented as the mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < as determined by one-way ANOVA followed by Tukey’s multiple comparison test.
(E) Primary human NK cells (n = 10 donors) were cultured for 6 days in the presence or absence of ALT-803 or IL-15 with or without IL-21 and with or without addition of IL-12 24 hr prior to harvest. IL-10 in the supernatant was quantified. Data are combined from three independent experiments and represented as the mean ± SEM; ND, not detected; ∗∗∗p < 0.001, ∗∗∗∗p < as determined by one-way ANOVA followed by Tukey’s multiple comparison test, comparing ALT-803 or IL-15 culture groups.
(F) Mice (n = 10/group) were infected with PbA then treated therapeutically with PBS, IL-15C, or ALT-803 on days 2 and 3 pi. Data are combined from two independent experiments.
Related to Figure S6.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions

9. Figure 7
IL-10 Expression by NK Cells Is Required for IL-15C-Mediated Survival from ECM
(A) Survival curve of WT or Il10−/− mice infected with PbA then left untreated or treated therapeutically with IL-15C on days 2 and 3 pi. Groups: PbA (WT), n = 15; PbA + IL-15C (WT), n = 15; PbA (Il10−/−), n = 7; PbA + IL-15C (Il10−/−), n = 13.
(B) Survival curve of mice infected with PbA then left untreated or treated with IL-15C on days 2 and 3 pi along with either a control Ab or anti-IL-10R Ab on days 1, 4, and 6 pi (n = 9–10 mice/group).
(C) 4 × 106 WT CD8+ T cells and 4 × 106 Il10rb−/− CD8+ T cells were transferred into congenically disparate WT mice on day −6, treated with IL-15C on days −5 and −3, and infected with PbA on day 0. Proportion of WT or Il10rb−/− CD8+ T cells in the spleen and brain on day 6 pi. ∗p = 0.01 as determined by 2-way ANOVA followed by Sidak’s multiple comparison test.
(D) Survival curve of mice infected with PbA then left untreated (n = 13), treated with IL-15C therapeutically (n = 14), or given 3 × 106 NK cells isolated from IL-15C-treated WT (n = 25) or Il10−/− mice (n = 21) on day 2 pi.
(A, B, D) ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < as determined by log-rank (Mantel-Cox) test. Related to Figure S7.
Immunity  , e4DOI: ( /j.immuni )
Copyright © 2018 Elsevier Inc. Terms and Conditions